Legal claims defining the scope of protection, as filed with the USPTO.
1. A parallel processing device comprising: a plurality of processors that execute respective processes and cause a field programmable gate array (FPGA) circuit to process part of the processes, wherein a first processor of the plurality of processors requests the FPGA circuit to execute processing of part of a first process currently being executed by the first processor, when the first processor transitions to an execution of a second process, the first processor saves environment setting information including data stored in internal registers of the first processor, sends movement-destination processor information including an identifier of a movement-destination processor to the FPGA circuit, and causes the movement-destination processor to obtain the saved environment setting information for continuous processing of the first process, and the movement-destination processor performs setting the environment setting information in the movement-destination processor during the execution of the processing of the part of the first process by the FPGA circuit, and continues the processing of the first process based on the environment setting information in response to a processing completion notification from the FPGA circuit.
2. The parallel processing device according to claim 1 , further comprising: the FPGA circuit that processes the part of the processes executed by at least one of the plurality of processors.
3. The parallel processing device according to claim 2 , wherein the FPGA circuit sends the processing completion notification to the movement-destination processor of the movement-destination processor information.
4. The parallel processing device according to claim 2 , further comprising: a plurality of processing completion notification signal lines that transmit the processing completion notification between the FPGA circuit and the plurality of processors, wherein the FPGA circuit transmits the processing completion notification via any of the plurality of the processing completion notification signal lines.
5. The parallel processing device according to claim 4 , wherein the first processor sends information on a request-source processor to the FPGA circuit when the first processor requests the FPGA circuit to execute processing of part of the process currently being executed, and the first processor transfers the processing completion notification from the FPGA circuit to the movement-destination processor when the first processor initiates the execution of the second process.
6. The parallel processing device according to claim 1 , wherein the first processor sends information on a request-source processor to the FPGA circuit when the first processor requests the FPGA circuit to execute the processing of the part of the first process, and transitions to a wait state of the processing completion of the FPGA circuit for a first predetermined time period.
7. The parallel processing device according to claim 6 , wherein the first processor does not initiate the execution of the second process during the first predetermined time period.
8. The parallel processing device according to claim 1 , wherein each of the plurality of processors repeatedly executes one process of the plurality of processes for a second predetermined time period and then executes another process for the second predetermined time period.
9. The parallel processing device according to claim 1 , wherein each of the plurality of processors activates and executes, when the processor receives an interrupt signal during execution of any process, another process having a priority higher than a priority of the process currently being executed, and stops the execution of the process currently being executed.
10. The parallel processing device according to claim 1 , wherein the movement-destination processor transitions to a wait state of the processing completion of the FPGA circuit after completion of the setting of the environment setting information.
11. The parallel processing device according to claim 1 , wherein the first processor sends information on the first process to the FPGA circuit when the first processor requests the FPGA to execute the processing of the part of the first process.
12. A method of a parallel process of a parallel processing device including a plurality of processors that execute respective processes and cause a field programmable gate array (FPGA) circuit to process part of the processes, the method comprising: requesting, by a first processor of the plurality of processors, the FPGA circuit to execute processing of part of a first process currently being executed by the first processor; when the first processor transitions to an execution of a second process after the requesting, saving, by the first processor, environment setting information including data stored in internal registers of the first processor, sending, by the first processor, movement-destination processor information including an identifier of a movement-destination processor to the FPGA circuit, and causing, by the first processor, the movement-destination processor to obtain the saved environment setting information for continuous processing of the first process; performing, by the movement-destination processor, setting the environment setting information in the movement-destination processor during the execution of the processing of the part of the first process by the FPGA circuit; and continuing, by the movement-destination processor, the processing of the first process based on the environment setting information in response to a processing completion notification from the FPGA circuit.
Unknown
September 11, 2018
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.